Roughly 37% of new dementia cases in older Americans may be linked to diseases that begin outside the brain, according to a massive Medicare analysis of 20.8 m beneficiaries. That figure, known as the population attributable fraction, estimates how many cases might be prevented or delayed if certain cardiometabolic conditions such as diabetes and hypertension were eliminated. The lead author of the Medicare study said the findings show that “treating common heart and metabolic diseases is not just about avoiding heart attacks, it is also about protecting the brain.”
The Key Study and Its Findings
The new analysis, described by the research team as a Primary nationwide assessment, used a 2017 Medicare dataset covering 20.8 m beneficiaries to look at how non-brain illnesses relate to later dementia diagnoses. Investigators focused on eight cardiometabolic diseases, including diabetes and atrial fibrillation, and tracked which beneficiaries went on to develop dementia in order to estimate how strongly each condition raised risk. The work, which has been summarized for clinicians in an EMJ neurology report, framed these disorders as potentially modifiable levers for reducing dementia at a population level.
Using standard epidemiologic methods, the authors calculated a combined weighted population attributable fraction, or PAF, for the eight cardiometabolic diseases and found that together they accounted for 37% of incident dementia cases. In other words, if those diseases could somehow be removed entirely, the model suggests that roughly 1 in 3 new dementia diagnoses in this Medicare group might not occur, although the authors stressed that PAF is an estimate rather than a direct measure of preventable cases. An Authoritative PAF analysis on the same dataset provides the DOI and PMID for verification and describes how each disease was weighted before being combined into the 37% figure.
How Non-Brain Diseases Reach the Brain
Cardiometabolic illnesses such as diabetes and hypertension affect blood vessels throughout the body, and researchers argue that the brain is no exception. Chronic high blood pressure can damage small cerebral arteries, while insulin resistance may disrupt how neurons use energy, creating a pathway from “body” disease to cognitive decline. In the Medicare analysis, the authors describe their approach as Primary prevention-focused and point to vascular inflammation as a plausible bridge between conditions like atrial fibrillation and later dementia, a view echoed in a Nature cardiometabolic study that links systemic metabolic stress to brain changes.
Some scientists go further and argue that infections starting outside the brain may directly seed neurological damage. A Primary periodontal study reported finding P. gingivalis DNA and toxic gingipain proteins in Alzheimer brain tissue, and the levels of these bacterial markers appeared to track with Alzheimer pathology. In mouse experiments within the same research program, oral infection with P. gingivalis led to brain colonization and Alzheimer-like damage, suggesting that a chronic gum infection might, in some circumstances, reach the brain and contribute to disease.
Broader Context from Global Research
The Medicare findings sit within a much larger effort to quantify how many dementia cases worldwide might be delayed or avoided by targeting modifiable risks. An Foundational modeling paper that looked at seven risk factors initially calculated a combined population-attributable risk of 49% under the assumption that each factor acted independently. When the same authors adjusted their model to account for overlap between risks such as low education and physical inactivity, the combined fraction dropped to about 28%, giving rise to the widely cited idea that roughly a third of dementia cases might be linked to preventable exposures.
More recently, an Authoritative Lancet Commission update expanded the list to 14 modifiable risk factors across the life course and estimated that up to 45% of dementia cases could be prevented or delayed by addressing them. A Major academic summary of the same Lancet Commission framework emphasizes that these figures reflect population-level shifts rather than guarantees for any one person, and that interventions need to start early and be sustained. Together, these global estimates align with the Medicare PAF of 37% for cardiometabolic diseases, suggesting that a substantial share of dementia burden links back to non-brain conditions that are already common targets of public health campaigns.
Government Baselines on Modifiable Risks
The U.S. Government has begun weaving dementia prevention into broader chronic disease guidance, reinforcing the idea that what harms the heart can also harm the brain. Official prevention materials from the Government Alzheimer program highlight high blood pressure, uncontrolled diabetes, physical inactivity, hearing loss, smoking, and excessive alcohol use as modifiable risks that appear to raise dementia likelihood. While these materials do not provide the 37% figure from the Medicare analysis, they frame dementia risk through the same cardiometabolic lens, noting that conditions such as high blood pressure and diabetes are more common in some racial and ethnic groups and may contribute to disparities in dementia rates.
These Government resources also stress that risk accumulates over decades, which matches the life-course approach of the Nature cardiometabolic work and the Commission and global estimates. While the Medicare study focuses on people aged 65 and older, the prevention message from public agencies is that managing blood pressure, blood sugar, and lifestyle earlier in life may be just as important for protecting cognition. In that sense, the 37% PAF for cardiometabolic diseases can be seen as the late-life expression of risks that have been building for years.
Infections, Immunity, and the Oral–Brain Connection
The periodontal findings have fueled a broader debate about whether some dementia cases might be partly infection-related. A Primary review of infection hypotheses describes how chronic pathogens, including P. gingivalis, herpesviruses, and others, might trigger sustained immune activation that gradually damages neurons and synapses. In addition to human brain studies, an experimental paper showed that P. gingivalis can persist inside human iPSC-derived neurons in vitro, with gingipain activity linked to Alzheimer-like changes such as tau disruption and synapse loss, findings that Adds mechanistic support to the oral infection theory.
Some clinicians remain cautious, noting that the presence of bacterial DNA in brain tissue does not prove that infection caused the disease, since severe dementia might itself make people more vulnerable to infections. A clinical overview from a major academic center on whether some Alzheimer cases are infection-related stresses that current evidence is largely correlational and that randomized trials of antimicrobial or anti-inflammatory strategies are still lacking. Even so, the periodontal work fits conceptually with the Medicare message that diseases and exposures outside the brain, including chronic oral inflammation, can have long-term consequences for cognition.
Why This Matters for Prevention
If cardiometabolic diseases account for a combined PAF of 37% in the Medicare population, then small improvements in how those conditions are prevented and treated could translate into meaningful reductions in dementia incidence. The While official prevention guidance already encourages control of blood pressure and diabetes, the new analysis suggests that neurologists, cardiologists, and primary care clinicians might need to frame those goals explicitly as brain-protective. A commentary on the Medicare findings in an Authoritative PMC record notes that cardiometabolic diseases cluster in communities with limited access to preventive care, which could mean that dementia prevention strategies must also address structural inequities.
Vaccination research provides another example of how non-brain interventions might lower dementia risk. A large observational study of older adults found that receiving the shingles vaccine was tied to significant reductions in dementia, heart disease, and stroke, with an approximate 20% drop in dementia risk among vaccinated individuals, according to an analysis summarized by CIDRAP vaccine researchers. A separate team at a Major academic center reported similar patterns in a Stanford shingles study, and a broader media review of vaccines and dementia risk described converging evidence that immune challenges outside the brain might influence long-term cognitive outcomes. While these studies cannot prove causation, they reinforce the idea that protecting overall health can have spillover benefits for the brain.
Lifestyle Signals and Everyday Choices
Beyond formal diagnoses, everyday habits also appear to shape dementia risk, often through the same cardiometabolic channels highlighted in the Medicare analysis. A feature on modifiable risks from a Major outlet covering the Commission and prevention agenda points to physical activity, diet quality, and social engagement as levers that influence blood pressure, weight, and vascular health. These factors, in turn, align with the Government emphasis on controlling high blood pressure and diabetes as part of dementia prevention.
Even common routines such as beverage choices may matter at the margins. An observational study reported that a daily coffee or tea habit was tied to a lower risk of dementia, although the authors cautioned that caffeine drinkers may differ in other health behaviors that were not fully captured. While such findings are not part of the 37% cardiometabolic PAF calculation, they highlight how multiple modest influences, from blood pressure control to diet and oral health, can add up across a lifetime.
Uncertainties and Limitations
Population attributable fractions are powerful for public health planning, but they are not the same as proving that eliminating a risk will directly prevent a given share of cases. The Foundational PAF model that saw its combined estimate fall from 49% to 28% after accounting for overlapping risks illustrates how sensitive these numbers are to assumptions. The Medicare cardiometabolic study faces similar constraints, since conditions like diabetes, hypertension, and atrial fibrillation often travel together and share upstream causes such as poverty, limited access to care, and structural discrimination that are not fully captured in administrative claims.
The Medicare dataset also covers people aged 65 and older in the United States, which raises questions about how well the 37% PAF generalizes to younger populations or to countries with different health systems. Researchers involved in the Authoritative global report have called for more longitudinal studies that track risk factors and cognition over decades, as well as randomized trials that test whether aggressively managing cardiometabolic disease can slow or prevent dementia. Infection-related theories face their own uncertainties: a review of whether Alzheimer may be infection-linked notes that periodontal and neuronal studies remain largely correlational, and that alternative explanations, such as reverse causation or shared vulnerability, have not been ruled out.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
